Plot topic per-term per-topic probabilities — plot_topic_term • TextAnalysisR

Plot per-term per-topic probabilities with highest word probabilities.

Usage

plot_topic_term(data, ncol = ncol, topic_names = NULL, ...)

Arguments

data: A tidy data frame that includes per-term per-topic probabilities (beta).
ncol: A number of columns in the facet plot.
topic_names: (Labeled) topic names
...: Further arguments passed to dplyr::group_by.

Value

A ggplot object output from stm::stm, tidytext::tidy, and ggplot2::ggplot. The result is a ggplot object representing the topic-term plot.

Examples

suppressWarnings({
if(requireNamespace("quanteda", "tidytext")){
dfm <- SpecialEduTech %>%
       preprocess_texts(text_field = "abstract") %>%
       quanteda::dfm()
data <- tidytext::tidy(stm_15, document_names = rownames(dfm), log = FALSE)
data %>% examine_top_terms(top_n = 2) %>%
plot_topic_term(ncol = 3)
}
})